Docs › Setup guides

Use SteadyRouter with Cline

Cline can talk to SteadyRouter in two ways. Use the Anthropic provider for Claude models, so prompt caching and extended thinking work the way they do with Anthropic's API, and use the OpenAI Compatible provider for every other model. You can set up both and switch between them in Cline's settings.

Before you start

Create a SteadyRouter account.
In the console, open API Keys and copy your key. New accounts come with one called Default Key, on Lowest price (see below). To make another, click Create API Key. Keep Smart routing on: if a group fails, the request is retried in the next group within the same call. A new key starts on Auto, or on Lowest price if your balance is under US$1.
Find the model name you want on Model Square. Model names have no vendor prefix, for example claude-sonnet-5-5 or gpt-5-mini.

Claude models: the Anthropic provider

Open Cline's settings (the ⚙️ icon in the Cline panel).
Set API Provider to Anthropic.
Paste your SteadyRouter key into Anthropic API Key.
Check Use custom base URL and enter:
URL
```
https://api.steadyrouter.com
```
Pick a Claude model under Model, for example claude-sonnet-5-5.

Requests go to SteadyRouter's native Anthropic endpoint (/v1/messages), which passes cache_control and thinking through as is. Calling Claude through the OpenAI-compatible format only suits plain chat, so use this setup for coding.

Other models: the OpenAI Compatible provider

Open Cline's settings (⚙️).
Set API Provider to OpenAI Compatible.
Base URL:
URL
```
https://api.steadyrouter.com/v1
```
OpenAI Compatible API Key: your SteadyRouter key.
Model ID: a model name from Model Square, for example gpt-5-mini.
Under Model Configuration, set Context Window Size and Max Output Tokens to match the model. Cline can't read these from a custom provider.

Trying it on a small balance

New accounts currently start with US$0.05 of credit. Before each request, the gateway reserves balance based on the input plus the maximum output the client asks for, and coding clients ask for a lot: often around 20,000 tokens of context and 32,000 or more output tokens. With a small balance:

Use Lowest price under Smart routing on your API key. Default Key is already on it; for another key, choose it when you create the key or later by editing it.
Start with a smaller model. On Lowest price, a request to claude-haiku-5-5 or gpt-5-mini costs a small fraction of the same request to larger models such as claude-sonnet-5-5, gpt-5.5 or claude-opus-5-5.
With the OpenAI Compatible provider, keep Max Output Tokens at 8192 or less. This lowers the amount reserved per request.

Once you've confirmed everything works, add funds to use larger models, and switch the key back to Auto if you prefer.

Lowest price often puts Pool groups first: usually cheaper, sometimes less stable. In every smart routing mode, Pool and Mixed groups can be used, so for stability-critical work, turn smart routing off and list only cloud or API groups on the key. See Models and groups.

If something goes wrong

What you see	What to do
404 when connecting	Check the base URL: `https://api.steadyrouter.com` for the Anthropic provider, `https://api.steadyrouter.com/v1` for OpenAI Compatible.
401 `Invalid token`	The key is wrong, disabled, expired or out of its spending cap. Check it under API Keys.
403 `insufficient_user_quota`	Your balance can't cover the request. Top up, or follow the small-balance steps above.
503 `model_not_found`	The model name is wrong, or the groups your key uses don't offer it. Check the name on Model Square.
408, 504 or 524	An upstream timeout. These are not moved to another group automatically, so retry.

Every error message ends with a request id. Include it when you email support@steadyrouter.com.

More: API docs · Model Square